Cocaine use has become an increasing concern over the past several years. The use of smoked cocaine has increased from 1% in 1977 to 18% in 1984 due to low cost and rapid onset of effects. More recently the use of smoked cocaine has declined among recreational users, however, the use among those addicted to cocaine remains high. In spite of this concern over the use of smoked cocaine, few studies have examined the effects of smoked cocaine in humans. Furthermore, little systematic research has been conducted on pharmacological and behavioral factors which may affect different stages of smoked-cocaine use. In part, the paucity of research in this area reflects the difficulty of administering precise doses of smoked cocaine which is necessary in parametric studies. Our laboratory has developed a method of delivering precise doses of smoked cocaine in humans. Over the past two years during the previous grant period, we have conducted several studies which have examined the effects of multiple doses of smoked cocaine in both humans and animals and factors which affect the reinforcing properties of cocaine. There are two major human experiments proposed for the current grant application. The results from these experiments will provide human laboratory models of self-administration. The goal of the first experiment is to develop a model of self-administration which incorporates a work requirement or a motivational aspect of self-administration. Subjects will be required to work on a task to earn tokens, valued at a fixed amount ($3.00) or a varied amount ($1, $3 or $5), which can be exchanged for various doses of cocaine (.1, .2 or .4 mg/kg) or a fixed dose of cocaine (.2 mg/kg) respectively. This model will be compared with the current self-administration model we have developed which does not include a work requirement. The goal of the second experiment is to investigate the effects of sertraline, a serotonin re-uptake blocker, on the self- administration of smoked cocaine. Toxicity of sertraline when administered in conjunction with cocaine will be determined first. Then the effects of sertraline on both models of self-administration (with and without work requirement) will be examined.